Exceptional responders in oncology: A systematic review and meta-analysis of patient level data

Purpose: We aim to systematically review and analyze the available literature on "exceptional responders" in oncology. We hypothesize that survival or patients with an exceptional response may be predicted based on clinical factors. Materials and Methods: A PICOS/PRISMA/MOOSE selection protocol was used to find studies that reported oncology patients with an exceptional response. A total of 333 initial articles were screened, and 76 articles were included, accounting for 85 patients. The primary outcome was survival after exceptional response therapy (ERT). The secondary outcome was survival since diagnosis. Univariate and multivariate analyses were conducted for both outcomes with 17 covariates. Results: The median age was 52 years (interquartile range, 35-66 y), 51.8% were male individuals, 18 (21.2%) had lung cancer, and 1 patient (1%) met all National Cancer Institute criteria for exceptional response. The most common treatment resulting in exceptional response was a form of chemotherapy (49.2%) followed by targeted therapy (26.8%) and radiation therapy (7.7%). The median time from diagnosis to initiation of ERT was 7.92 months (interquartile range, 0-24.72 mo). On multivariate analysis of survival after initiation of ERT, there were no predictors of exceptional response. On multivariate analysis of survival since diagnosis, predictors of prolonged survival included time between diagnosis and ERT initiation (hazard ratio, 0.52; 95% confidence interval, 0.32-0.87; P=0.0124) and single prior surgery versus none (0.08; 95% confidence interval, 0.01-0.98; P=0.04853). Conclusions: There were no clinically apparent patient or treatment factors that predicted favorable survival following ERT; instead, reporting of exceptional response appears to be biased.